c_cpp/lib/klib/bgzf.h

   1 /* The MIT License
   2
   3    Copyright (c) 2008 Broad Institute / Massachusetts Institute of Technology
   4                  2011 Attractive Chaos <attractor@live.co.uk>
   5
   6    Permission is hereby granted, free of charge, to any person obtaining a copy
   7    of this software and associated documentation files (the "Software"), to deal
   8    in the Software without restriction, including without limitation the rights
   9    to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
  10    copies of the Software, and to permit persons to whom the Software is
  11    furnished to do so, subject to the following conditions:
  12
  13    The above copyright notice and this permission notice shall be included in
  14    all copies or substantial portions of the Software.
  15
  16    THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  17    IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  18    FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
  19    AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  20    LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
  21    OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
  22    THE SOFTWARE.
  23 */
  24
  25 /* The BGZF library was originally written by Bob Handsaker from the Broad
  26  * Institute. It was later improved by the SAMtools developers. */
  27
  28 #ifndef __BGZF_H
  29 #define __BGZF_H
  30
  31 #include <stdint.h>
  32 #include <stdio.h>
  33 #include <zlib.h>
  34
  35 #define BGZF_BLOCK_SIZE     0x10000
  36 #define BGZF_MAX_BLOCK_SIZE 0x10000
  37
  38 #define BGZF_ERR_ZLIB   1
  39 #define BGZF_ERR_HEADER 2
  40 #define BGZF_ERR_IO     4
  41 #define BGZF_ERR_MISUSE 8
  42
  43 typedef struct {
  44     int open_mode:8, compress_level:8, errcode:16;
  45         int cache_size;
  46     int block_length, block_offset;
  47     int64_t block_address;
  48     void *uncompressed_block, *compressed_block;
  49         void *cache; // a pointer to a hash table
  50         void *fp; // actual file handler; FILE* on writing; FILE* or knetFile* on reading
  51 } BGZF;
  52
  53 #ifndef KSTRING_T
  54 #define KSTRING_T kstring_t
  55 typedef struct __kstring_t {
  56         size_t l, m;
  57         char *s;
  58 } kstring_t;
  59 #endif
  60
  61 #ifdef __cplusplus
  62 extern "C" {
  63 #endif
  64
  65         /******************
  66          * Basic routines *
  67          ******************/
  68
  69         /**
  70          * Open an existing file descriptor for reading or writing.
  71          *
  72          * @param fd    file descriptor
  73          * @param mode  mode matching /[rwu0-9]+/: 'r' for reading, 'w' for writing and a digit specifies
  74          *              the zlib compression level; if both 'r' and 'w' are present, 'w' is ignored.
  75          * @return      BGZF file handler; 0 on error
  76          */
  77         BGZF* bgzf_dopen(int fd, const char *mode);
  78
  79         #define bgzf_fdopen(fd, mode) bgzf_dopen((fd), (mode)) // for backward compatibility
  80
  81         /**
  82          * Open the specified file for reading or writing.
  83          */
  84         BGZF* bgzf_open(const char* path, const char *mode);
  85
  86         /**
  87          * Close the BGZF and free all associated resources.
  88          *
  89          * @param fp    BGZF file handler
  90          * @return      0 on success and -1 on error
  91          */
  92         int bgzf_close(BGZF *fp);
  93
  94         /**
  95          * Read up to _length_ bytes from the file storing into _data_.
  96          *
  97          * @param fp     BGZF file handler
  98          * @param data   data array to read into
  99          * @param length size of data to read
 100          * @return       number of bytes actually read; 0 on end-of-file and -1 on error
 101          */
 102         ssize_t bgzf_read(BGZF *fp, void *data, ssize_t length);
 103
 104         /**
 105          * Write _length_ bytes from _data_ to the file.
 106          *
 107          * @param fp     BGZF file handler
 108          * @param data   data array to write
 109          * @param length size of data to write
 110          * @return       number of bytes actually written; -1 on error
 111          */
 112         ssize_t bgzf_write(BGZF *fp, const void *data, ssize_t length);
 113
 114         /**
 115          * Write the data in the buffer to the file.
 116          */
 117         int bgzf_flush(BGZF *fp);
 118
 119         /**
 120          * Return a virtual file pointer to the current location in the file.
 121          * No interpetation of the value should be made, other than a subsequent
 122          * call to bgzf_seek can be used to position the file at the same point.
 123          * Return value is non-negative on success.
 124          */
 125         #define bgzf_tell(fp) ((fp->block_address << 16) | (fp->block_offset & 0xFFFF))
 126
 127         /**
 128          * Set the file to read from the location specified by _pos_.
 129          *
 130          * @param fp     BGZF file handler
 131          * @param pos    virtual file offset returned by bgzf_tell()
 132          * @param whence must be SEEK_SET
 133          * @return       0 on success and -1 on error
 134          */
 135         int64_t bgzf_seek(BGZF *fp, int64_t pos, int whence);
 136
 137         /**
 138          * Check if the BGZF end-of-file (EOF) marker is present
 139          *
 140          * @param fp    BGZF file handler opened for reading
 141          * @return      1 if EOF is present; 0 if not or on I/O error
 142          */
 143         int bgzf_check_EOF(BGZF *fp);
 144
 145         /**
 146          * Check if a file is in the BGZF format
 147          *
 148          * @param fn    file name
 149          * @return      1 if _fn_ is BGZF; 0 if not or on I/O error
 150          */
 151          int bgzf_is_bgzf(const char *fn);
 152
 153         /*********************
 154          * Advanced routines *
 155          *********************/
 156
 157         /**
 158          * Set the cache size. Only effective when compiled with -DBGZF_CACHE.
 159          *
 160          * @param fp    BGZF file handler
 161          * @param size  size of cache in bytes; 0 to disable caching (default)
 162          */
 163         void bgzf_set_cache_size(BGZF *fp, int size);
 164
 165         /**
 166          * Flush the file if the remaining buffer size is smaller than _size_
 167          */
 168         int bgzf_flush_try(BGZF *fp, ssize_t size);
 169
 170         /**
 171          * Read one byte from a BGZF file. It is faster than bgzf_read()
 172          * @param fp     BGZF file handler
 173          * @return       byte read; -1 on end-of-file or error
 174          */
 175         int bgzf_getc(BGZF *fp);
 176
 177         /**
 178          * Read one line from a BGZF file. It is faster than bgzf_getc()
 179          *
 180          * @param fp     BGZF file handler
 181          * @param delim  delimitor
 182          * @param str    string to write to; must be initialized
 183          * @return       length of the string; 0 on end-of-file; negative on error
 184          */
 185         int bgzf_getline(BGZF *fp, int delim, kstring_t *str);
 186
 187         /**
 188          * Read the next BGZF block.
 189          */
 190         int bgzf_read_block(BGZF *fp);
 191
 192 #ifdef __cplusplus
 193 }
 194 #endif
 195
 196 #endif